Blindsight & the bakery facade

This Post provides a link to Chapter 9 of “What Scientists can Learn from Artists”, the book that presents the science that supports so many of the ideas and proposals found in my other three books. Its full title is “Blindsight, unilateral neglect and the bakery facade illusion”. It recounts what for me was a particularly exciting adventure into the mysteries of eye/brain function. This had its origins in research I was doing at the University of Stirling, Scotland, and reached its conclusion here in Montmiral, as a result of trying to understand why a student, who was good at drawing accurately, persisted in seeing the slope of a wall top differently from three other people sitting close to him. It turned out to be a question of his being taller than them and, as a result, he was relating the wall top to a slightly different background, with fascinating consequences: Ones which were to provide the substance not only of the chapter to which this Post is linked, but also of a closely related chapter in my book on drawing.*

Piazza del Duomo, Milan, which plays an important role in a fascinating, game changing story told in this chapter.

Two paradoxes

The bringing together information about “blindsight”, “unilateral neglect” and “The bakery facade illusion” provides yet another approach to making  clear that the process of “seeing” is complex. It also presents evidence that lead to two paradoxes, namely that:

  • we can all “see” what we cannot see
  • we can all “imagine” what we cannot imagine”.

Luckily knowledge of eye/brain systems can make sense of these seemingly senseless propositions. Also, it can alert artists to some deep seated problems they cannot avoid facing when drawing or painting from observation.



As I usually do when presenting book chapters, I am providing below an edited version of the “Introductory” to the chapter in question in the hope that it will whet your appetite for reading the chapter itself.


This chapter delves a little deeper into the subject of visual mechanisms and systems. It is one of the most important in the book because it provides information concerning the central problem as to how preconscious, bottom-up processes enable top-down control of the skilled use of eye/hand coordination. The first part takes the form of a detective story. The key to unlocking the mystery lies hidden in two experiments, relating to two visual impairment syndromes, each resulting from damage to a different part of the brain. Though other syndromes can be legitimately given the same names, they will be referred to as “blindsight” and “unilateral-neglect”. The second part of the chapter describes a powerful visual illusion, first noticed in relation to the facade of a building in Castelnau de Montmiral, S.W. France. This is shown to have general implications both for artists trying to depict scenes containing rectangular surfaces and for psychologists of perception, trying to understand the mechanisms underlying analytic-looking.


WHAT SCIENTISTS CAN LEARN FROM ARTISTS” CHAPTER 10 – “Blindsight, unilateral neglect and the bakery facade illusion”


*  “Axes of symmetry, recession and the constancies”: A chapter from my book on drawing that shows some of the ways the theory in this more scientific chapter can be related to practice.


Other chapters from “What Scientists can Learn from Artists”

Like the chapter to which this Post is linked, the links below can be used to access chapters from the middle section of my  book that elaborates on the science behind subjects that feature in the other volumes:

Go to top of page

Go to list of all other contents


19 thoughts on “Blindsight & the bakery facade”

    1. It is fascinating what science has been finding out about the nature of “seeing”. The plot thickens in the chapter that I will be posting next week. It tells about the astonishing (well it astonished me and all who were with me when we saw it for the first time) demonstration of Johannson that makes it quite clear how unimportant outlines (contours) are to recognition of moving objects. Perhaps we should not be surprised for “recognition” means seeing something as being the same on more than one occasion, and, as Degas and common sense have told us, outlines, even of the same solid, static object, perceived on two occasions, at marginally different distances or angles, are never quite the same (except in the unlikely event that both object and viewer stay in exactly the same relationship to one another). It wouldn’t be much use to us if we could only recognise things that we are seeing from exactly the same distance and viewing angle as on previous occasions. It follows that the processes that enable “recognition” cannot give much priority to outlines, which is why most people can navigate their lives perfectly well without paying much attention to them. No wonder I encounter so many students who find learning to draw accurately such a challenging struggle.

  1. Francis, this is intriguing and also quite difficult. I’m still not sure how Lovis Corinth did his self portrait. He can’t have used the same method as the man with his back to the Duomo in Milan, so how can he have produced it?

    1. Good question. One of my students, Keith Williams, worked with a unilateral neglect patient. He explained that this man had told him that he experiences a fleeting impression of a whole scene immediately before his visual world divides into two halves. However, the sense of an integrated whole is ephemeral, for it only lasted a matter of milliseconds. As soon as he focuses down on a target object, only half of it can be seen. The selection of the object to be looked at and the actions required for analysing it would have been predetermined by the preconscious use, first of recognition systems and then of action-organising ones. In other words, the choice of what to look at and how to interact with it had already been made.

      The first step in the visual part of any interaction with the target object will always involve focusing down, and it is doing so that necessitates the putting in place of an axis of approximate symmetry. In the case of unilateral neglect patients, this step is accompanied by the loss of half the scene they are looking at. Keith described how frustrating his charge found this outcome and how hard he struggled to reconstitute the fleeting sense of wholeness.

      My first thought was “why can’t the unilateral neglect patients just swivel their heads until the targeted object moves into the centre of the visible half of the scene?”. Later I was told that other researchers had had the same thought and that head swivelling was indeed the basis of the treatment of unilateral neglect.

      The good news is that this strategy can work. The bad news is that it can take a long time to do so. I cannot remember if I was informed about typical recovery times, but I am fairly sure that Lovis Corinth had to struggle on for many months or even years, before he could give the impression that he was painting normally again.

  2. I will always remember the excercise of the bakery roof … How sad it is not to see these lines anymore ! along the years, I saw many many good paintings, drawings of that view, Sarah’s works for exemple , astonishing !

    1. Just to explain to readers who do not know. The bakery building was converted and the characteristics which led to it being the source of the “bakery facade illusion” were removed.

  3. When I worked with the unilateral neglect gentleman, it was many years ago. Teaching art in a day centre, the gentleman wanted to draw. When he drew a house, it became clear, the stroke he had, damaged a specific part of the brain, as he only drew a little more than the right half of any object. Each time I asked him to draw what he could see; and when working from like, all he drew was delicate and well considered/observed. The question that interested me and I could not answer, was the point when unilateral neglect actually happened. In looking at a landscape, or even a room, if you only see half, it is still a landscape or a room. Is unilateral neglect happening all of the time, but only an issue when focussing on an object? Had I asked him to draw a landscape, would he have drawn a full landscape? I don’t know, because that experiment was not attempted, but I feel he would still only create half a drawing. If a landscape filled his whole understanding/perception, at what point does it become half a drawing? My gentleman seemed to have no difficulty in understanding the world around him. The issue was not so much his sight, but how it was interpreted. Many years later, I still do not understand where in the process the actual unilateral neglect takes place, but it must be somewhere within the processing of information, as that is where the stroke occurred and damage took place.

    It did not have to be direct focus. For some experiments, I asked him to draw a house from memory and when he ‘pictured’ a house in his mind, the unilateral neglect was present. Such images were never as considered, as when he drew from life.

    A few years earlier, when working at another day centre, a colleague worked with a gentleman having unilateral neglect; and my first experience of the condition. She taught needlework/cross-stitch. He would copy complex magazine/calendar images, drawn out onto canvas, but only managed the right half, or just over. Like my gentleman, the image was not split equally between left and right, as the worked area progressed a little beyond halfway, but not reaching three-quarters. In that area, there seemed to be some understanding, but not full, as the lines and marks faded; there was never an abrupt junction. Some lines, like the roof of a house, my gentleman extended a little further than halfway, but not windows etc. He seemed to know the building went further, but where were those lines drawn from?

    For both, the unilateral neglect only concerned the left side, which indicates a stroke on the right side of the brain. And for it to be very localised, as there are many strokes and very few who suffer unilateral neglect. Both also had limited use of their left arms.

    My colleague’s remedy was to turn the image and needlework upside-down, and the gentleman completed his work; not possible in my experiment using a vase of flowers. His speech was also very limited, but he could shout a few expletives when frustrated. My gentleman could communicate his frustrations, albeit with a struggle. I was sure, both men understood something was not right, compounded when the piece they had been working on was turned upside-down and they could see their deficiency. I then worked with my gentleman to clarify what had happened, which eased the frustration as medical staff at the time were unable to explain.

    1. Wonderful to hear from you Keith. What you told me about unilateral neglect when you were here at the Painting School put another piece n the jigsaw of my understanding of the analytic looking cycle, and you might be surprised to know how often I remember you with gratitude. I am sure that anyone who read my Post will be really interested in your comment. Thanks so much.

  4. What I did not mention, was when the gentleman walked, he felt fine when in the open with the whole environment around him, but walking close to something like a wall, he was not sure what was there. There seemed nothing wrong with his vision, but the perception of what he saw. I assume walking in the open, everything was whole. At what point, did the whole become left and right?

    It would be interesting for scientists to look deeper into this condition, as to where the unilateral neglect takes place, and why. I am also interested in the overlap, where perception ends. Both gentlemen could draw/cross-stitch beyond the half-way point, but then their work faded to nothing. Was it from perception, or memory of what should be there? That element still amazes me and I do not have any answer. Is there an overlap with perception like stereo vision?

    Having read your chapter and comments, writing my comment took me back to a very enjoyable time in my life, which through time I had lost, until I needed to think about it again.

    1. What you write is very interesting. The evidence I have gathered together suggests that the the point at which a sense of the whole becomes halved is the moment when the analytic looking systems click into action. The process of recognition is preconscious, so has nothing to do with our sense of what we see. Its purpose is to key us into instructions as to how to react to whatever has been recognised. It is important to emphasise that recognition is not just a matter of identifying objects but also of keying into instructions as how to react to them (for example, how to locate the handle, pick them up and drink from them). If this involves finding further information about the object (for example, the location and shape of the handle) the reaction will require analytic looking and contain instructions as to how to proceed with the organisation of actions. It is at this point that the constancies if ‘size’, ‘orientation’ and ‘shape’ are required and applied. It is consistent with what you told me long ago and with what you write now, that this is also when people with unilateral neglect loose their sense of wholeness and find themselves able to see only the right hand side of matters.

      In the next Post on movement – – you will find evidence from Gibson, Bernstein and Johannson that we don’t rely on analytic looking to guide ourselves around the world (although we can use it to help in certain situations). Your evidence makes clear that the unilateral neglect patient that you describe has no problem using whole-field information when being guided by the systems Gibson describes (based on information provided by what he calls ‘flow fields’). For example these can guide him down a pathway that leads to a wall and tell him when he is getting near to the wall. If he carries on, at some point, they will inform him that, if he does not adjust matters, he will bang into the wall. It is at this point of decision that his analytic looking systems are likely to click in. It is when they do that he looses half his visual field.

      Keith I am so grateful for this information which fills in my understanding of these issues. I would be very grateful if you would tell me whether, in your view, what I have written fits in with your very clearly presented observations.

      1. Firstly, I am a painter not a scientist. Throughout I have accepted and used in my work, the science that helps me paint better. Working with this gentleman was to aid your scientific understanding. In doing so, I also gained an understanding, which led to a few of my own questions, as I raised in my last post, but there was no real desire to search for answers. I feel scientists should find them; to me, at this stage, it is a scientific issue, rather than artistic.

        From working with this gentleman, rather than the one my colleague worked with, I gained a better understanding of his ‘whole-field’ and ‘object related’ issues. The other gentleman had a more intense stroke that also restricted his walking and speech. I would say your explanation seems valid. From what I experienced I could not argue against; but I am not a scientist, so I do not feel qualified to say it is 100% right, only that it fits with my observations and understanding.

        There is a point, when the whole becomes an object, which is an important factor with unilateral neglect. But it was not reliant on vision, as the same happened from imagination when he drew a house using his memory. Therefore, it has to be connected with perception in the brain. Whenever he needed to consider the ‘object’, in being a wall, or vase of flowers; or even the house from his memory, then unilateral neglect became an issue. The action his brain was being asked to undertake, has to be key in pinpointing the root. He could find his way around the day centre and carry out other activities unaided. There should have been other experiments, or observations that were not carried out. Each day he attended, a lunch was provided. I don’t know if looking at a plate of food, he only ate from the right side. My feeling is, someone prompted him, or rotated his plate. There were so many things I could have established, but I only worked with him for a couple of weeks and I wanted to complete the research quickly to forward my findings; because of the painting school, drawing seemed the most important element.

        I felt it was important to explain the drawings/cross-stich in detail, especially their ability to continue a little beyond mid-point, and then the perception faded away. Without that fact, it could be assumed there was a sharp division at the mid-point. I consider that ‘overlap’ to be important in fully understanding perception. As we see in stereo, with an overlap in both eyes, I do wonder if perception also operates in stereo. Are there two receptors, that consider left and right, the left one being damaged by the stroke? That left receptor would also reach a little beyond the mid-point, as the right seems to. In my mind, it would make sense. Humans understand their environment better with stereo vision, the same may be the case for perception. Why have stereo vision, if you cannot perceive in stereo?

        I do feel, as you have written about, unilateral neglect could be compensated for; but the issues against are numerous.

        Unilateral neglect is usually created by a stroke and only in a very small percentage. It does not occur often enough to be seen as a major issue.
        The strength of any stroke is the most important issue. Having a minor stroke, where the blood clot happened in the exact place to cause unilateral neglect, could be ‘treated’ easier, like walking and speech, which can also return quickly.
        Strokes are more likely to happen in advanced years, when there is less time to compensate for the condition; as well as added complications.
        People need to be diagnosed with the condition, which I do not believe is being assessed or understood by ‘professionals’ even today; it certainly was not in the early 1990s. (Although my gentleman was aware of an issue, no one else had noticed, or explained what was happening, it took an artistic activity to recognise the issue, then science to help him understand. An artist may be aware of an issue as soon as he/she tried to draw. Or maybe not, as the drawing is still an object, of which, the person might only perceive the right half; though from my experience, there would be an understanding, if only that something was not ‘right’.)
        Speed of diagnosis is important, as the greatest improvement occurs in the first few months following a stroke; speech, mobility and even unilateral neglect, but relies on therapists being available and an understanding of the condition.
        Other issues may seem more important to the patient and/or others, like speech and walking.
        Finally, there is the type of person, some may not want to know, or do anything about the condition.

        1. In addition: They are only my observations, mainly of one person. This subject needs more research to gain a clearer picture. If there were ten, or more people with unilateral neglect, observed and assessed over a longer period, concrete facts could be offered. From what I achieved, I am not able to say, my information is consistent for all, as it is not proven, though feel it could well be.

          I would like that group of people to have their vision tested. Can the person see fully, partially, or not at all through their left eye? How does it compare with their right eye? Has their vision changed since the last time it was tested? For me, the test would include more than reading letters, as they are object-related, not only each letter, but also the board they sit on. It would need to be a combination of ‘environment’ and ‘object-related’ imagery. How many flowers do you see in the vase? How many cows do you see in the landscape? These days, with headsets and 360 degree projections, ‘real’ situations could be replicated. Look into each eye, to see if there is any damage. The issue to me, is not vision, so these tests I would want as confirmation.

          The same tests could also assess perception. By moving, rotating, even multiplying objects, people with unilateral neglect could see/perceive different things. When and where do changes occur in perception? Is there a left-side deficiency in both eyes? Where is the point when our brain recognises an object, near or far distance, or depending on the object? There is so much potential for the right person(s). But that is not me! Finding ten people with unilateral neglect, capable and prepared to participate would be difficult, even spread over a number of countries and years. I was very fortunate to find someone so capable and keen to participate.

          My client knew there was a wall to his left when walking along a corridor, but when next to it, he could not see/perceive the wall. As he started to walk down the corridor, the wall in going off into the distance, would have been part of the whole picture, so he knew and remembered it was there. In walking, combined with understanding, he could walk along, but held his fingers out a little to feel if the wall was too close. Also, being able to accurately see and assess the right wall, he could walk parallel with it; and when unable to detect the left wall, he naturally adjusted other sensors to guide him. Able to think logically, he quickly made decisions based on the established as a whole, but also repetition, in walking from dining area to the art room, at the other end of the building twice each day he visited. When walking along the corridor, had he looked directly at the wall, some of it at least, would have been visible/perceptible. Had he turned around, all the previously left wall, would be on the right and visible.

          When my gentleman, drew beyond the mid-point, he may have moved his head slightly to perceive from a different angle, thereby gaining a little more information; but the same applied when he drew from imagination. Did he imagine changing positions? If so, he could well have imagined more, than the same amount from observing directly. Each time, I witnessed him drawing and did not notice any perceptible movement.

          When drawing from life, I always asked my students to draw/paint only what they saw. I worked with many people having poor eyesight, even a young gentleman who was blind, yet wanted to draw, nothing was impossible. Although on times, the gentleman with unilateral neglect was frustrated with the ‘gap’ created by his limitation, my expectations, understanding and explanation helped. Each time he drew for me, it was a positive experience for both of us.

          You write: ‘It is important to emphasise that recognition is not just a matter of identifying objects but also of keying into instructions as how to react to them (for example, how to locate the handle, pick them up and drink from them).’ If you cannot see an object, there is no chance of understanding, other than from memory. Even if you can see an object, there are times when you may not perceive/understand. Your comment made me think of some other clients/patients. More recently, while working in hospitals, I occasionally came across older people with advanced dementia. Some could not identify/understand a mug (with handle) of fluid placed in front of them, yet claimed they wanted to drink. Even prompted, they could not recognise the mug, or what to do with it. If the mug was lifted, they could follow the object moving. When someone helped by raising the mug to their lips, natural reaction took over and it was drunk. I assumed the dementia had damaged the receptors in total, not just one side; but they seemed able to see and follow an object, it was not the same as unilateral neglect. If unilateral neglect was present on both sides, would the person be able to perceive any objects, or even the whole? With so little capacity to understand, it was impossible to establish anything relevant to science. For me, there are too many unanswered questions and why proper scientific tests need to be carried out.

          1. Thanks so much Keith. Everything you write is interesting. Clearly, the patients are lucky to have someone so thoughtful and observant to look after them.

  5. I stopped teaching within those environments around 1997, finding the demands impacted negatively on my own work, which was developing at a pace; and still does. For me, it has been important to keep work and painting separate.

  6. A fascinating and very detailed investigation [and conversation in the comments section] about visual perception and its impact on drawing strategies. Sadly it seems that very little of this is ever incorporated into Fine Art courses such as degree courses. Only at Montmiral!

    1. If you search online for blindsight/unilateral neglect, there are plenty of sites that give a general ‘scientific’ explanation, even demonstrating the issues through drawing, but the thrust seems to be helping individuals how to compensate.
      In my response, I questioned if we use perception in stereo because we see, and therefore should understand in stereo; but I only worked with individuals having left-side neglect. Reading further online, the condition can occur on the right side as well. There were images of men, only shaving exactly half of their beard. From my experience, the exact division I find strange, as my clients could perceive a little more than half-way, which is why I though perception had to be a stereo function. Having written that comment, I now know it is in stereo. One experiment, was to draw the face of a clock. With both eyes open, most of the numbers, 1 to 12, were positioned around the right half of the clock. With eyes closed, the results were much better, almost fully accurate. The error level, I felt, was no different to anyone attempting the same exercise with their eyes closed. It would seem we perceive differently when our eyes are closed. Why? Where is that division? I would like more research into perception and the differences between eyesight and brain. My tests were visual, but the condition goes much deeper than visual perception, as it impacts on other senses, certainly touch, but does that also include hearing, smell and taste?
      So few encounter blindsight/unilateral neglect, who are young enough and capable enough to participate in research. Both of the people I knew had strokes late in life, but it is not the only cause. Often with a stroke, the damage also takes speech and limb movement, which to the individual, even some ‘professionals’, those areas can seem more important to regain, or compensate for. It is a rare condition, that has required time to build sufficient understanding for a general explanation. Even now, 25 years after I carried out those experiments, we can make ‘artistic’ statements about perception, but nothing has been formalised within science. Nothing has been drilled down far enough to become fact.
      How much of the science actually helps you to physically paint better? There needs to be more research undertaken, specifically the impact on painting/drawing. The majority of art students, accept the science that enables them to paint/draw better. I fall into that category; I am a painter, not a scientist. When you can interpret that knowledge into practical terms, it may feature within courses materials and mass-market books. Until then, it will remain in the realms of science for only a minority to find.

Leave a Reply

Your email address will not be published. Required fields are marked *